298 RECENT PROGRESS OF THE THEORY OF VISION. 



by practice to recognise the parts of which it is com- 

 posed, and to separate them from one another. The 

 former is our usual (though not constant) habit when 

 looking with one eye, while we are more inclined to the 

 latter when using both. But inasmuch as this incli- 

 nation must chiefly depend upon practice in observing 

 distinctions, gained by preceding observation, it is easy 

 to understand how great individual peculiarities may 

 arise. 



If we carefully observe the rivalry which ensues when 

 we try to combine two stereoscopic drawings, one of which 

 is in black lines on a white ground and the other in 

 white lines on black, we shall see that the white and 

 black lines which affect nearly corresponding points of each 

 retina always remain visible side by side an effect which 

 of course implies that the white and black grounds are 

 also visible. By this means the brilliant surface, which 

 seems to shine like black lead, makes a much more stable 

 impression than that produced under the operation of 

 retinal rivalry by entirely different drawings. If we 

 cover the lower half of the white figure on a black ground 

 with a sheet of printed paper, the upper half of the com- 

 bined stereoscopic image shows the phenomenon of Lustre, 

 while in the lower we see Retinal Rivalry between the 

 black lines of the figure and the black marks of the 

 type. As long as the observer attends to the solid form 

 of the object represented, the black and white outlines 

 of the two stereoscopic drawings carry on in common the 

 point of exact vision as it moves along them, and the 

 effect can only be kept up by continuing to follow both. 

 He must steadily keep his attention upon both drawings, 

 and then the impression of each will be equally combined. 

 There is no better way of preserving the combined effect 

 of two stereoscopic pictures than this. Indeed it is 

 possible to combine (at least partially and for a short 



